Settings that control how a model generates output: temperature adjusts randomness/creativity (low values = more deterministic, high values = more creative); top-p (nucleus sampling) limits selection to the tokens whose cumulative probability reaches a threshold; top-k limits selection to the k most probable tokens.
In AIF-C01 (Domain 3) and GENAI-L (Domain 3) you're tested on the effect of each parameter. Typical question: reduce the variability of a technical support chatbot's responses to make it more consistent → lower the temperature.
Inference parameters (temperature, top-p, top-k) is a concept that can show up on the exam. Check the full guide to see it in context.